ML20154D647

Forwards Insp Rept 50-275/70-05 on 701201-03.No Nonconformance Noted

ML20154D647
Person / Time
Site:	Diablo Canyon, 05000000
Issue date:	12/29/1970
From:	Spencer G US ATOMIC ENERGY COMMISSION (AEC)
To:	James O'Reilly PACIFIC GAS & ELECTRIC CO.
Shared Package
ML20154C370	List: ML20154C367 ML20154D294 ML20154D425 ML20154D452 ML20154D528 ML20154D575 ML20154D594 ML20154D647 ML20154D736 ML20154D846 ML20154D855 ML20154D869 ML20154E066 ML20154E143 ML20154E188 ML20154E240 ML20154E258 ML20154E356 ML20154E392 ML20154E431 ML20154E484 ML20154E507 ML20154E577 ML20154E597 ML20154E600 ML20154E634 ML20154E666 ML20154E684 ML20154E762 ML20154E785 ML20154E809 ML20154E821 ML20154E871 ML20154E878 ML20154E908 ML20154E958 ML20154E967 ML20154E977 ML20154F001 ML20154F024 ML20154F073 ML20154F094 ML20154F104 ML20154F116 ML20154F123 ML20154F135 ML20154F145 ML20154F153 ML20154F156 ML20154F390
References
FOIA-88-156 NUDOCS 8805190279
Download: ML20154D647 (2)

Text

-

( .

l

,? 4. UNITED STATES p r, ' [.

ATOMIC ENERGY COMMISSION O ,, ,

1 DIVISION OF COMPLI ANCE RCGION V '

allt BANcROFT WAY BERKELEY. CALIFORNI A 94704  % . wi.aisi arr. es

> umber 29, 1970 I

J. P. O'Reilly, Chief Reactor Inspection and Enforcement Branch Division of Compliance, Headquarters l

PACIPIC CAS AND ELECTRIC COFJANY - DIABLO CANYON UNIT NO.1

i. DOCKET NO. 50-275 The attached report contains the details of our recent inspection of construction activities at the site of the subject facility. No items of' i

nonconformance were identified. The announced inspection was conducted on December 1, 2 and 3,1970 pursuant to PI 3800/2 and in accordance with the master inspection schedele for the project.

The PG&E QA investigation concerning oue previous observation of possible' inadequate technique in performing liquid penetrant tests appcored to be '

iimely and comprehensive. The results of the investigation failed to show a general deficiency in the PDM QC pron com. Dosed on our review of the l investigation effort, we considered PG6E's followup action concerning the issue to be adequate.

PC&E's in test findings concerning the voll thickness of the spools of primary piping in storage indicates possibly that the Westinghouse and Vendor QC inspection efforts may not be sufficiently comprehensivo in scope.

We concur with PG6E in that more extensive examinations of the piping need to be performed before valid conclusions can be reached. We intend to pursue 1

the subject and will report fully on PC&E's additional test efforts and conclusions after the next scheduled inspection (March 1971).

g i

Generally, our inspector found adequate documentation concerning construction discrepancies and QA audits. In particular, the documentation of the * {

occurrences involving the steam generators and the reactor vessel showed  ;

the complete history of relevant facts, circumstances, evaluations and actions. 1

\

. You will note that the removal of the eco water from the generators 'was effective and that an article quoted in the report indicates that the exposure to sea water under such conditions does not create chloride stress corrosion problems.

' The inspection of the reactor vessel penetration tubes (af ter being struck by a wooden panel) was considered cociprehensive and should have detected any damage. Also included in the report are data obtained from

• PC&E's independent tests which demonstrate the detrimental effect of arc burns on reinforcing steel since it may be of interest to other inspectors.-

8905190279 080510 .

IL a-156 PDR

( ( .

4 J. P. O'Reilly December 29, 1970 d...

.. i

, The events associated with the discrepant anchor bolts accentuated the li need for implementation of a formal procedure clearly delegating specific

!l responsibilities to appropriato personnel to confirm that the informacion

shown on certifications is consistent with what the certification implics.

. The PC&E on-site QC personnel began this practice after our Juno inspection.

l i

G . S .- pencer j - Senior Reactor Inspector '

i

Enclosure:

l CO Report No. 50-275/70-5

by A. D. Johnson ,

l' ded. 12/29/70 '

l j ec: E. G. Case, DRS )

P. A. Morris, DRL **' l' i l

R. S . Boyd, DRL (2) ,

'. R. C. DeYoung, DRL (2) ,

P. W. Howe, DRL (2) . l A. Giambusso, CO '

i L. Kornblith, Jr. , CO R. H. Engelken CO Regional Directors, CO '

j REG Files ,

1 J

0 -

S a 1 l -

i l

+ .

a 8 k t

4

- - - , , . , - - , - , , -e .----. - -

l

( ( .

l 1

' '~ '

U. S. ATOMIC ENERGY COMMISSION -

i '

p,.,

i

, DIVISION OF COMPLIANCE 1 REGION V '

(

il ,.

p . , i .

E' Report of Inspection 1.g .

CO Report No. 50-275/70-5 ,

il .

I d Licensee Pacific Cas & Elcetric. Company j Diablo Canyon Unit No.1 i

' ' Construction Permit No. CPPR-39 i Category A i.

, Date of Inspection: -

December 1, 2 and 3, 1970 l '

Date of Previous Inspection: September 15 and 16, 1970 i -

Inspected by: 'f1 CMC-k '

~

t.s A. D. Johnson Date Reactor Inspector

{

I Q .

Reviewed by: -

M . #fMCdw /4/;7"/7s

, G. S. Spen'cer Date i

, Senior Reactor Inspector 1 j . '

Proprietary Information: None i '

,

• SCOPE Tvoe of Facility:

Pressurized Water Reactor -

Power Level 3250 net

. Locattent

Diablo Canyon, San Luis Obispo County, l California Tvoe of Inspectioni '

Routine - Announced

., Accompanyinz Personnel:

R. W. Smith, Director and C. S. Spencer Senior Reactor Inspector accompanied i the inspector on a tour of the site and related facilities on December 3,1970.

Scooe of Inspection:

Review (1) status of previously identifi !

[,7 items of concern, (2) discrepancy report.

hd-7 f /8/= V [O) (3) QA oudit activities.

( (

l 2- .

. . 9*

i . '

' S U} MARY i ..

1 Sa f ety It ems _ - None 1 1

Nonconformonee Items _ - None j Status of Previousiv Reported Items - No safety or nonconformance items were I

reported in the previous inspection report.

Other Significant Teems -

t

, 1., overall completion of construction activities was estimated to I be 18.7% on December 1,1970. (Section 3.)

2.. Ultrasonic inspection of primary piping spool pieces in storage i indicated that one identification stamping had breached the

- minimum design wall thickness. Also random checks of the wall thickness identified a small section of wall area on one spool, piece that was less than the minimum design thickness. More '

.- tests are planned to' be performed by Westinghouse and PG&E to determine the accuracy of the UT cost results. (S ection C. I . ) I '.  ;

1

'3. An evaluation of the constructed mockup section of the area between the wide beam flanges and the containment liner indicated that concrete can be placed in accordance with the planned procedures

.vithout detrimoncal effects to the quality of the end product.

, (Section C.2.)

4 Plans are to have a procedure which will delineate the specific

, responsibilities (within PG&E) for verifying that all certification documents are valid and proper, approved and implemented in the immediate future (within 60 days). (Section C.3.)

5. A review of QA discrepancy reports indicated that disposition of items continued to be accomplished pursuant to the PG&E QA standard procedure. (Section D.)

6. Anchor bolts installed in the containment base net for the steam generater supports were found by PG&E to be in nonconformance ,

with the specified ASTM standard. Since the design of the supports had not been completed the designer changed the design to accomodate use of the installed bolts. Other bolts not conforming to the specifications were rejected. (Section E.)

7. Sea water leaked into three of the steam generators during

transporattion from Tampa, Florida to the site. The generators have been cleaned and resealed for storage at the site. (Section F.)

i

, . ,, ,_ - , . . . . . - - - ,, n- e n

1

( ,

.j .

I i4 **

18. During transfer activities associated with the reactor vessel a wooden panel collided with several of the instrument penetration, tubes. No damage resulted. (Section G.)

l

9. The on-site QA engineers have been implementing an active audit program. (Section H.)

10. PG6E's QA section has determined that liquid penetrant tests are being performed pursuant to the appropriate procedures. (S ection C.4.) -

l

11. ' Tests performed by PG&E demonstrated that arc burns on reinforcing steel may be detrimental to the. physical characteristics.

(S ection D.3.)

Mananement Interviev The inspector met with Messr. Hersey, Friedrichs, Hickran, Richards 'and {

other members of PG&E's on-site staff. The scope and findings of the '

1 inspection were reviewed. - I Mr. Hickman stated that in view of their findings concerning the wall  !

thickness of the piping spool pieces in storage, additional tests will be performed on all of the primary piping in storage to assure design and code requirements are satisfied.

Mr. Friedrichs stated that their evaluation of the concrete mockup section has assured PG6E that the quality of concrete placed between the wide beam flanges and the containment liner is adequate.

Mr. Hersey stated that from QA audit results and their experience associated I with the anchor bolts, all contractors and PG&E construction staff have been emde aware of the importance of promptly submitting to PG&E a formal report of all discrepancies identified. He also said PG6E QC personnel are to check appropriate certification documents of materials used by contractors at the site to assure the material conforms to the specified standards.

• The inspector commented that the QA followup investigation concerning the previously observed questionable liquid penetrant testing appeared to be comprehensive and satisfactory to indicate that the contractor is following the prescribed procedures. .

DETAILS A, Persons Contacted '

H. R. Hersey -

Project Superintendent .

R. R. Friedrichs -

Resident Civil Engineer '

A, W Hickman -

Resident Mechanical Engineer

. e e

. . _. . ~ . _ _ . -. - _ _ _. . .

4- -

p*.'

E . G. V. Richards ,- Director, Quality Engineering - San Francisco I n F. W. Brady -

QA Engineer k P. L. Bussolini -

QC Coordinator l QA Engineer I R. W. Wood -

i iS B. Good -

QA Enginect - San Francisco i' L. J. Garvin -

QC Engineer - San Francisco E

1 B. Construction Status 1

- Overall completion of constru'ction of the Diablo No.1 project was 2

estimated by PG&E's Construction Department to be approximately 18.77. on i December 1, 1970. Specifically, the containment tuilding was shown to bc

397, complete with 347 of the liner installed. Construction of the auxiliary

, building vas considered to be opproximately 48 percent complete. Unofficial

projections indicate hot functional tests will not commence until sometime after January 1973 and probably not until March 1973 or later.

C. Resolution of Previous Issues i

During the previous visits several items had been brought to the .,

licensee's attention for evaluation and possible action. The following '

information summarizes the disposition or status of the itema.

1. Indentation Stampinn of Cinnn I pipinn t '

! PG&E's inspection finding showed that identification markings on j

the primary loop piping in storage at the Pismo Beach facilities

)* were not in compliance with the Westinghouse specifications.

Additional tests have been performed to determine whether the impression depths found to be outside of the Westinghouse speci-f fications infringed on the minimum design wall thickness of the

{ pipe. During November 1970, ultrasonic test equipment was used to measure the wall thickness of the pipe. The test results showed that eight indentations associated with one identification merking had infringed on the minimum design vall thickness. A, test of the section of the piping free frca stamping was also check ed . The results showed that an area of about ten square

' inches was slightly less than the minimum design wall thickness.

As e result of PG&E's test data Westinghouse personnel were scheduled to arrive on site during the week of December 7,1970 with special optical measuring equipment. This equipment was stated to supposedly be abic to moosure the wall thicknesses within an accuracy of one mil. Mr. Hickman stated that as a result of their findings they intend to do more testing of the other spools of pipe to assure thenselves that the other scetions aset the minimum design requirements. Hickman showed the inspector

4

(

C g 4**

• results of measurements made by Westinghouse of the wall thickness

' prior to shipping the spool pieces to the site. Each spool piece was measured at locations of approximately one inch and two feet from the ends of each pipe. These measurements indicated that all wall thicknesses were within the minimum design limits. Hickman indicated to the inspector that he believed the ultrasonic test technique used by PG&E had cn accuracy value of within at 1 cast 10 mils and was probably accurate to within 2 or 3 mils. It appeared from the (PG&E) UT test data that the maximum infringement on the minimum design thickness was approximately .050 inches.

2. Placement of Special Concrete in Area Between Wide Beam Flannes -

To assure that the concrete placed between the wide beam flanges and the containment liner at the base of the containment wall would be of proper quality, a mockup section was poured and was

, later inspected by the PG6E Construction Engineers. (See C0 Report No. 50-275/70-4) Af ter the normal curing period, the forms were stripped from the mockup. The concrete was thoroughly examined and no voids or other imperfections were found. The inspector,,

examined the mockup section and concurred with the PG&E findings.

3.. Verification of Ouolity Control Information on Certification Documents Mr. Richards indicated that a procedure to assign specific responsibilities pertaining to the verification of quality control information on certification documents would be formalized and approved within 60 days. Mr. Dussolini, QC coordinator, said that the General Construction Department personnel at the site had been reviewing all material certifications relating to materials used in the civil structures. Richards indicated that formulation of the proposed procedure had been delayed because of differing views concerning who exactly should be responsible for review of the various certifications. According to Richerds, this especially applies for the itema independently purchased by PG&E from outside vendors. .

4 Adec'>acy of Dvo Penetrant Tests During the previous inspection, Mr. William Kelley, Construction Inspector, Region II, observed a Pittsburgh-DesMoines Secc1 Co.

inspector performing a dye pencerant examination on one of the liquid holdup tanks. Kelley felt that the inspector did not remove enough dye penetrant during the cleaning phase of the test and as a result there was a pinkish cast of the developer af ter it vos applied. He also noticed that the developer was applied 9

gs.

t k

f ,

from an aerosal can in high winds and the inspector was shielding g the work with his body in such a way that he could not see how much developer was applied.

Mr. Brady, site Quality Assurance Engineer, investigated the facts and circumstances surrounding the criticisms voiced by Mr. Kelley.

Brady's report of his investigation stated that it was difficult to assess the criticism of PDM's procedures. According to Brody, both L. J. Garvin and W. Wood, who were with Kelley when he watched the test, agree that the developer appeared pink in spots. Mr.

Garvin, who is qualified as ASNT Level II for dye penetront examinations, felt that the test was adequato for detecting defects that would be rejected under ASM2,Section VII or ASF2,Section III.

Class C. The criteria for dye penetrant removal is by its naturo, not very specific. ASME,Section VIII, which is the governing codo for this test, says ..."ony pencerant reemining on the surface shall I be removed. Insufficient removal will leave a background which will interfere with subsequent indication of defcces. Care must be exercised that penetrant in any defects is not removed." This seems to leave room for judgement as to when just enough penetront in removed. Similarily, there are no hard, fast rules for applying the developer.

The test in question was performed on a particular completed wold seam af ter grinding. The liquid holdup tanks involved are fabrice-ted from t-inch to 3/16-inch stainlese secel. The seems era ucided from two sides, typically with rwo or three passes on one side, followed by grinding the exposed side of the root pass, and then completed with rwo or three passes. The dye penetrant test is performed after grinding the root pass in order to locate any defects before the weld is completed. This allows correction with a minimum amount of repair work.

PDM has been advised, by letter, of the concern over their procedures.

In answering, they observe that the cleaning phase is difficult because the work is in cho wold groove. However, they reinstructed their inspectors to exercise additional care in obtaining complete excess penetrant removal.

Mr. Brady indicated that he had personally watched three PDM dye penetrant examinations since the previous inspection. Two of these were in a weld groovo. The procedures were noted to be 1cteer

, perfect. In all of the tests, the inspector removed the pencerant by first rubbing with dry cloth and then rubbing with a cloth slightly dampened with cicaning solvent. The developer was sprayed on in one coat and any thin spots were sprayed with a second coat.

k e

____w

, , . , , ~ . , . - _ . . - - - - -. 4 -, --- r_ 7

I l

i

\

{

1 g..

E During the test in a weld groove Mr. Brady noticed a light pink ,

color where the grinding marks were close together. However..he I Y: stated that he thought this was inherent in testing a ground

'}

surface rather than a result of imperfect technique, i'

y

In conclusion Mr. Brady indicated that he could not draw any i definite enclusions about the test that Mr. Kc11cy witnessed. I t

There seemed to be agreement that the developer had a light pink j color. It could be that this was careless procedure. It could l also be a reflection of the difficulty encountered in performing j

. auch a test in a wcld groove. Or it could be a combination of the  ;

' two. In any case. Mr. Brody concluded that the PDM and PG&E '

personnel were qualified for the jobs that they were doing, and {

that the PDM dye penetrant procedures they were now using vere acceptable.

, l

, D. Construction Discrepancies +

1 1

~ l

1. , Deviations i

' The inspector reviewed eightoon deviation reports which havo been prepared since the previous inspection. Although several of tho l items have not been fully reviewed and resolved, the inspector

[ confirmed that the items vero being processed in accordance with

} the QA discrepancy procedure. The more significant items included:

i a. Sea water had 1 caked into the steam generator during transpor-

tation to the site. (Sce Section F. of this report for additionc !

i 8 detail.)

i I

b. A wooden panel collided with the bottom instrumentation '

penetration tubes of the reactor vessel while unloading it  !

from the barge at Avila Deach. (See Section G of this report l

for additional detail.) t

c. .. Penetration alignment was inconsistant with contract specificact f Design Engineer approved change so that alignment would be I consistent with requirements of Section VIII of the ASME Boiler and Pressure Vessel Code.

I

d. A penetration arrived at the site with defective welds.

Field repairs were made and the vendor was directed to improvo inspection efforts to preclude the shipping of defective materials.

, e. A penetration sleeve was found slightly out of-round. The Design Engineer accepted the sleeve as was since the slight out of-roundness vould not create future fitup problems.

1 I

l .

]

8-

z.  :

1, .

ir

-i U f. Several liner places were found to exceed the plumb specifi- l

!) , ,

cation in the contract. The Design Engineer accepted the l ji work as completed since his findings indicated the deviations had no effect on the engineering requirements, y, .

j h g. The refueling canal penceration was installed in an improper i

'j

, orientation;. The penetration was removed and was then properly I 3 i installed.

I.

] h. During rail transportation, the pressurizer vessel evidently l received a strong jolt. The recording of the accolcromoter l 1 monitoring the vessel had on one occasion exceeded the scale reading. Also a brace wcld had partially ruptured. In

! addition, during W journey the rail car had a "hot box" i i

which charred the wooden supports under the pressurizer. l Examination of the particular portion of the vessel adjacent i

to the wooden support showed no evidence of paint damage.

The paint is an aluminium type and should show damage if it vere exposed to excessive heat. Mr. Hickman believed that the vessel was undamaged by the shock since checks of the i , ,

l hesters in the ' pressurizer showed no defective conditions. l At the timo of the current inspection k'estinghouse had not l as yet determined the severity of the possibly shock shown i by the acceleromoter since the calibration data concerning

the monitor had not yet crtived at the site.

, 1. Receival inspection of one of the safety injection pumps indicated a metal to metal noise when the pump was rotated.

Plans were to inspect, repair and use if possible,

j. Receival inspection of the containment spray pump, serial No. 2213175-2, disclosed a crack in the pump casing near the <

bearing at the drive end of the pump. Plans were to repair l and use if possible, i k.

A 1007 review of the secom generator and other anchor bolts I by PG&E disclosed that all of the bolts for one reason or another failed to meec the ASTM specificatione. (See Section i I of this report for decatis.) .

2 Hinor Variations The inspector confirmed from o review of the minor variation log that the recorded items had been appropriately resolved at the

., field level. One interesting minor variation involved are burns 8

4 G

(

r . .

9

W-f on reinforcing steel. To determine the effect and consequences F of are burns on the integrity of the reinforcing steel, tensile

! i tests were performed on samples containing are burns. Significant f information from a report of the tests follows.

'I "Four (4) test sempics of reinforcing stoel, sizo 18, crode 60, vero procured for testing, Sample #1 vos the reference i standard to determine approximate values for yield strength,

! tensile strength and clongation since the heat number vos unknown. Sample #2 had an arc burn produced on it by dragging a weld rod across the robar. Sample #3 had an arc burn created

! by touching the rebar with a weld rod and pulling it off.

Sample #4 had an are burn created by holding a weld rod in one spot on the robar for a count of five (5), creating a

, . molten puddle three-cinhths (3/8's) of an inch in diameter.

~

The results are tabulated in Table 1 ,

i ,

TABLE 1 1 SAMPLE YLD. STRENGTH TEN. S EENGTH ELONG.. FAILURE ,.

1. 1 - Ref. 63,500 psi 102,500 psi 13% At Lower Jaw

1. 2 - Arc Drag 63,500 pai 103,000 psi A* l " 38" 15.1% Away From Buen

1. 3 - Spot 64,250 psi 103.750 psi 12.7% AhjLyy$mDu$n y

1. 4 - 3/8" puddle 64,000 psi 75,500 psi 1.8% Thru Are Burn The failure of Sample 04 occurred through the molten puddle L created by the are burn. It con be determined from the results i

that the are burn did indeed cause brittleness. A stress point

[ was created in the crea of the are burn causing failure and very little elongation or ductility. Sample #3 though satis-

' , , factory, gave evidence of a slight crack in the spot created by she arc burn.

'

• This test is by no means a method of determining the degree

' of are burns that might be acceptable in the field. It was performed solely to demonstrate the fact that are burne 6n reinforcing steel can be detrimental."

Richards indicated that since the probicm was identified, about eight defects have been repaired by installing a codweld splico.

,, He also said that contractor personnel are exercising more care to preclude are burns.

( (

6***

b E. Other Class I Structures (Steam Generator Supportsi b

h As a result of the inspection con ucted during June 1970, the inspector

[ reported in CO Report 50-275/70-3 that the mill certifications for the anchor bolts installed in the containment base slab showed the bolts to be

, of proper quality. This conclunion was based on a sample review of the ASTM A-490 Bole certifications. Also, this was the some timo that the inspector raised the question as to whether PC&E QC personnel were revicwing u all documented information for conformance to code requirements. Subsequent 1 to the inspectors departure, the on-site QA staff performed a completo

. review of all the bolt certifications. The audit findings were as shown i in Appendix A.

! The Deficiency Report submitted to PC6E by PDM stated that the REC 1

Corporation received material for the anchor bolts from Ryerson and Crucible Steel Companies along with mill test reports. Apparently the person that has the responsibility for checking the mill test reports was on vacation.

A substitute who was not totally qualified approved the mill test reports as conforming to the ASTM A490 specifications. The REC Corporation proceeded I

1 with the manufacture of the anchor bolts. .

I The mill test reports were submitted to Pittsburgh-Des Moines Stect Company's Purchasing Department which has an in-house Quality Assurance man assigned specifically to check mill test reports and materials being received, i The PDM Purchasing Department Quality Assuranco person noted the discrepancy

, and notified the REC Corporation that the tuterial was not in conformance vich ASTM A490. The REC Corporation notified PDM's Purchasing that the bolts were in specification and they would send proper mill test reports to ahov this. In the meantime, the correspondence involved took considerabic time and the REC Corporation completed the manufacture of the anchor bolts and shipped them directly to the job site.

On receiving the corrected forms, a discrepancy was again found since they did not show the physical proporties but had given hardnces equivalent resdiage. The Purchasing Department again went back to the REC Corporation and notified them that this was not adequate. PDM was told that the bolts vers adequate and REC would take care of the situation. Because of this ensuing difficulty and paper work, the bolts were installed in concrete prior j

to the correct paper work and mill test reports reaching the job site, ,

i As soon as it was discovered by the (PDM) Corporate Quality Assuranco Department that the above had occurred, immediate steps were taken to find out the full details of the problem and also to survey the REC Corporation in order to determins if they had the capability of producing the above enchor bolts.

2 i

e 4

0 4

- - , , , _ _ . , _ . . , . . . , - - - - , , - . . _ - . . -- , e - - - - . - , ,- -

C C

[ .

5 g.

s .

11 t

f f

I 1. PDM Quality Assurance conducted an audit of the REC Corporation

{ manufacturing procedures, quality assurance, and quality control organization on August 14, 1970 and found them to be satisfactory j to produce materials for this contract. PDM felt the action they had taken was satisfactory to see that this could not occur on i future orders.

E.

(

, 2 In addition to the above steps, PDM Quality Assurance will make cercain that all noterials and certifications are within the

contractual requirements of the specification prior to permitting j -

shipment of any future materials, from REC Corporation.

i In view of the above problems encountered with the anchor bolts, PDM i requested specification deviation approval that the anchor bolts, (32-11AR2,'

{ 64-11AR4, and 32-11AR6) which are now embedded in the concrete be lef t in j position, and that PG6E consider alternate designs as may be required to i accommodate the slightly lower physical properties of the embedded anchor

] bolts. In addition to the above stated ASTM A490 anchor bolts, there were anchor bolts, (32-11AR1, 64-11AR3, 32-11ARS, and 34-11AR7), that are above the floor embedment liner places. These anchor bolts will be reordered as soon as possible and will meet the ASTM A490 requirements, according to PDM.

Since the detailed design of tho supports had not been completed, PC&E's Design Engineer incorporated the "os installed" bolts into the support design and PG6E accepted the installed bolts, however, the remaining bolts not embedded in concrete were rejected.

F. Steam Generators Sea water entered three of the four steam generators via channel nozzles during their transfer from Tampa Florido to Avila Beach, California where the generators were unloaded from the borge and transported overland to the j site at Diablo Canyon. The condition was discovered during inspection ca 4

arrival at the site in July 1970 -

,- Cleaning of the generators was completed during October 1970. Efforts j to clean the generators were delayed during September due to labor disputes. '

A report of a meeting between Westinghouse and PG6E personnol discussed i

che design of the shipping closure of the secan generator channel nozzlos l

and why they failed. (Appendix B shows a diagram of the probicm crea.) The nozzle veld preparations were protected by an oncle iron ring scal-welded

' to the end of the nozzle outside of the weld preparations. Another ring was tack velded to this ring, inside it and the weld preparations, for further l

l i

l l

l

{ (

. .- 12 d**

• protection during manufacture. When the steam generators were prepared

' for shipment, the tack welded inner rings were to have been removed and

. gasketed covers bolted to the seal welded rings. However, the inner rings

" were not removed, and the covers were prevented from seating as intended, resulting in leakage paths past the tack welded joints. Plastic coatings were sprayed on the outside of the bolted closures, but the coating failed.

There was no procedure requiring removal of the temporary rings, although there was a note calling for their removal on a drawing. The procedure ha's now been revised to call for this step. Westinghouse stated to PG6E that they had such confidence in the closure design that they conducted no tests of the integrity of the shipping preparation.

Examination of the steam generators at the site showed the following:

1. 1. Water in nozzles on both inlet and outlet sides, estimated at 3 to 5 gallons, no evidence of water on the tube sheet.

1. 2. Water in nozzles and in head, ascimated at 5 gallons.

I I ' #3. Closure appeared to be intact. .

1. 4 Slight indication of water at ring of one nozzle.

In an effort to determine whether salt water entered any tubes, distilled water has been flushed thru the lowest tube of generator #2. ,

Analysis of this flushing water showed chlorides less than 1/10 ppm which )

indicated little or no sea water entered the tube. Westinghouse felt there is no need to flush other tubes in the generator. Discussica revealed that this procedure would have detected one cubic centimeter of sea water in the tube but would not have detected one drop of sea water in the tube.

According to the report there have been two other incidents of voter entering steam generators during shipment. The Indian Point Unit 3 generators had "an order of magnitude" more sea water enter than the Diablo generators.

The Zion (Unit 17) generators had condensation resulting fro ~m air leakage, but not sea water. These steam generators for the three projects were sh'ipped over a period of months. The failure of the closures was not detected until af ter all three units' generators had been shipped. In all cases the temporary shop protective rings had not been removed from the nozzles.

Steam generator tubes are inconel. Tube sheets are forged carbon steel with inconel explosivo cladding. Channot heads are cast carbon steel

. (SA216WCC) with steinless sceci (308L) weld clod. Channel splitter plates are inconel. The channel nozzle wcld preparations are weld built up of 308L

., over a transition pass of 309L. The hcod esseing is stress relieved afwer cladding but before the splitter place is welded in. The head to tube sheet l l

% l

_ . - - . . _ _ , _ .,_. -_- _.-_. - . __---.-.,,1

( ( .

'l ,c. '

il N weld is stress relieved by local heating. The splitter place welds are

'l not strer,s relieved.

i

!. The nozzle welds are not sensitized as defined by ASTM ?93 cr ASTM 262

i This is the so-called Strauss test and was indicated to be acceptablo by the AEC to determine sensitization. The test was made on duplicate material, not l on material from the nozzlea themscives.

I Inceael was considered to be immune to chloride ion stress corrosion cracking, but will pit on long exposure to sea water, especially if stagnant.

No pitting has been seen in the steam generators, which have had a relatively d short exposure.

-)

The corrective action proposed by Westinghouse was to vosh the channels with deionized water until a one ppm chloride level was reached in the wash I water and then swab with alcohel to dry. They believed that this would i

pruent any future damage from the sco water and was all that vos required.

j It was the same procedure proposed for the Indian Point generators which had greater leakage. Mr. Baulig, Westinghouse, said that 1/10 ppm chlorido was

' ,I relative *iy easy to achieve in the ficid, and it vsa his stated intention!to I

wash to this level, not just to the 1 ppm level required by . .np s .

PG&E accepted the investigation of the incident and the proposed cleaning procedure. General Construction prepared a quality control fiation report on this incident.

Westinghouse has changed the nozzle closure design from bolted to scal velded. Al Hickman PG&E Resident Mochonical Engineer stated a proforence for the original design, if correctly executed, because access for inspection 4

in the field is easier.

O The report of chloride analysis supplied by Finolysis Inc., Poso Robles, California, indicated that all final rinses had chloride concentrations of

.l {

0.1 ppa or less with the exception of two samples which showed concastrations 1

'; of 0.6 and O'.7 ppa. Of interest, m . Good supplied the inspector a copy of 4 l

on article titled Application oE Steels for Desolination Plant by Richard l T. Jones published in August 1967 in Volume 7 No. 3. Metals Engineering 1

Quarterly (ASM). The aritcle primarily deals with corrosion of stainless l

, steel exposed to salt water environments. The test durations were up to two i

,; years, i

The article indicated that to determine the effect of stress on the

'l corrosion resistance of the stainicas sccels (type 304 & 316). Ericsen

cups were formed by depressing a hordened steel ball into each stainless steel sample. Conclusions from examination of the stress cups under low l magnification and metallographically, indicated the absence of evidence of stress corrosion cracking or any preferential pitting corrosion on the cups.  !

i (A copy ,of the article is available in the Region V office.)

, , _..,,_,_____m.-__.__.., , _ _ _ . , _ . . . . _ . _ _ _ , _ _ , _ _ , , . _ _ _ _ _ , . _ . _ . ~ , _ ._-.,_,_._.,_,,,,._.,y, , ~ . _ _ _ _ ,

( ,

(

14 l

.P-U G. Reactor Vessel N The reactor vessel arrived at Avila Beach for unloading during September

'! 1970. During enloading operations Digge, the equipment handling contractor,

!' had pr viously removed.the wooden panci covering the 0-Ring box on the borge l,'i fore-deck and relocated the 0-Ring box to a safe area. While removing the wooden panel (10' x 22') that had been under tho 0-Ring box a 15 to 20 mph h; wind caught the panel and whipped it into the starboard mid-section of the

.! reactor vessel botton, dragging its edge across several of the instrument il penetration tubes. After an initial inspection, a detailed inspection

{ procecure formulated by Wesi.inghouse and PG&E was impicmented. Results of

the inspection were reported in the deviation report as follows.

'l i 1. The area of impact was determined by holes and abrasions in the protective cover. This area, in turn, was found on the reactor

vessel bottom by bits of wood on two (2) of the penceration tube

,, end expandable plug bolt throods. These two tubes had been marked ti initially because thef r plugs had been knocked loose and 1 caked y gas. A third tuba had its plug end bolt bent, but wasn't looking.

The fabric cover was removed during this process.

Tubes, S/N 456-09-2 and 456-15-2, had wood chips on their plus j bolt threads and tube S/N 456-212-2 had a bent bolt.

$ 2, Th'e burlap boots and polyethylene socks were removed from these three tubes and nine (9) other tubes adjacent to them. They woro

':I all checked for straightness by laying a straight-edge on two (2) c sides of each tube, 90 apart. No bent tubes were found.

'i The socks and boots were reinstalled with new tuck-tape.

3. All instrument penetration tube end expandable rubber plugs were

': checked for internal gas icokoge with an "Ultra-Sonic Translator" (leak detector). One (1) plug was found to be leaking and was

]i:

o re-seated and.re-tightened. The leak was stopped.

4. Each tube end plug was physically checked for seating, tightness y (soundness) and possible damage to the tube.

1,oose burlap boots were removed and replaced solidly with new

,; cape. No damage was found. .

-l S. On completion of the inspcetion, the protective f abric weather cover was replaced over the Reactor Vessel bottom.

l. The resulta of this inspection disclosed no physical damage to the bottom penetration tubes.

1 t

i a

n.

'e

(

pi Mr. Brady indicated that the vessel would be transferred to the site from Avila Beach during the week of December 7,1970. The inspector observed during a tour of the temporary storage area at Avila Beach that (1) the vessel was under surveillance by a guard (2) the vessel was covered with plastic and (3) the gas pressure was monitored by an installed gauge (0.5 to 1 psig of nirrogen is being maintained on the vessel).

H. OA Audits The on-site QA audit program was reviewed and discussed with Mr. Brady.

The review included a reading of recent audit reports, audit schedules and plans. Mr. Brady indicated that to perform on audit associated with a particular phase of contract work approximately 15 man days are expended in preparing, conducting and documenting a particular QA audit.

The audit reports appeared to be comprehensive in describing the scope and findings of the audits. Copics of the audit reports had been distributed to all interested parties and written responses of adverso findings were noted to be filed with the appropriate oudit report. Mr. Brady indicated that the need for timely filing of deviation reports had been highlichted.on several occasions. However, this probicm seems to have been remedied as contractors have become more aware of the urgency for prompt submittal of formal notification whenever discroponcies are identified.

Since July 1, 1970, 15 QA audits of construction activitics have been completed, plans were to complete three more audits before the end of the year pursuant to the approved schedule.

P Activities audited included: ,

1. QC inspections.

2. Transporting, handling and storage of steam generators and reactor vessel.

3. - Receival and storage 'of electrical equipment. ~

4 Installation of auxiliary salt water piping. ,

5. Mechanical equipment - inspection, storage and placement.

6. Fabrication of liquid holdup tanks.

7. Containment structure. '

a. Rebar

b. Liner

c. Wide Beam Flanges s

4

, - . , - - , - - . . - . - - - . - - . . = . . . . - , -

l

. 1 ASTM A-490_ - Specifications pi

" - 2 " diameter

' ~

gnimum Moxtrnum Tensile 150 k 180 k Yield 130 k Elongation in 2" 147.

Reduction in area . 40%

Hardness - Brinell 3.02 3.52 ANC110R BOLTS Puno supports - 64 bolts %" >

l 11AR1 & 11AR2 ,

Tensile 141,675 .

5.6% low Yield - 124,815 ,. 4.0% low Elongation 19.37. ok P.ed, in Area 60.6% ok Hardness 2.86 5.3% low Generater Supports - 128 bolts - 2" 11AR3 6 11AR4 .

Tensile 139,460 6.9% low Yield 121,798 6.2% low l Elongstion 28.7% ok Red, in Area 64.5% ok l

Hardness 2.69 10.9%

i

! Lateral Truss - 64 bolts - 2"

! 11AR5 & 11AR6 i '

i Tensile 158,750 ok i Yield 146,000 ok '

i Elongation 8.07. 43% l l Red. in Area 37.67. 6.07. Ick l i Hardness ,

311/321 ok l

Annulus Supports - 34 bolts k" 11AR7 l Tensile 161,000 ok

. Yield 150,000 ok Elongstion 1 07. 28.67. Iow

.. Red, in Area 36 .07. 10% low i

Hardness 321/331 ok e

S AnEsoix - A .'

A i

(' . ,, , _ .

(

LIQV10 ENVELOPE A

su LEAK AREA ru ,,. ~x ..m je i s ~

i,.

i 37////////////////// ///////i V//iTA> 1 5 /Nh//; // M PROTECTIVE N0ZZLE COVER 4

i RING ASSY.  !

i.  :

,TACK WELD (3 PLACES) N 1 .

?

i T.EMPORARY. INNER RING .

[ '

$, SE AL WE'"* I

{ t

!)  :

, GASKET

/ ,

f .

! b

NOZZf.E WELO PREP.

(

, a ,

- A

, t

, 1

...'.O. ' 'i,'! .,'.,'. ,,fl .

THE SALT WATER INTRUSION OCCURED BECAUSE:' , . . . ' ' '

. ,1-THE SHOP FAILEO TO REMOVE THE TEMPORARY INNER, l . ,. i RING PRIOR TO SHIPMENT OF STEAM GENERATORS.

2- THE TACK-WELDS PROJECTED SUFFICIENTLY TO PREVENT .

t .

..THE SHIPPING COVERS FROM SEATING ON THEIR GASKETS

)

3- THE "Ll0VIO ENVELOPE" FA! LED OURING THE BARGE TRIP i ALLOWING SEA WATER TO CONTACT THE METAL SURFACES.

APPENDIX - B GM167027 ATTACHMENT "B" DVR No.35 I wonovco or 1 i Q-gu . i i I I i I I i f

. i I ' ' -i i i i i ,

w,i ,1 c a ce t oa't  !

os ic e.ev.o = , e, , n, ,c,

,,o W' L"

' DIABLO C ANYON PROJECT UNIT ONE 7' ' * ' ' $ ' " '

6 l

cw Rm,3 s c .,

STEAM GENERATOR REACTOR COOLANT '

• lCll%S ,,,

mx ','- a N0ZZLE SHIPPING COVER DETAIL

. uno. . .. u .

ctNemat. cosistaveries oceam vcst o . . . . . = v . . . . v%

t

. _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ -